Consider the Maxwell kinetic Lagrangian


where the field strength is $F_{\mu\nu}=\partial_\mu A_\nu-\partial_\nu A_\mu$. A vector quantity can contain at most 4 degrees of freedom (d.o.f.). Just from the definition of $F_{\mu\nu}$ it is clear that the longitudinal degree of freedom $\partial_\mu\chi$ in the vector

$$A_\mu=A_\mu^\perp+\partial_\mu \chi$$

cancels out of $L_{kin}$. Additionally, one can use the gauge invariance $A_\mu\to A_\mu+\partial_\mu\xi$ to e.g. eliminate the spatially longitudinal component in $A_i=A_i^T+\partial_i A^L$. This shows that the massless Maxwell field only propagates 2 degrees of freedom.

However, if we add a mass term to the Lagrangian:

$$L_m=-\frac{1}{2}m^2A_\mu A^\mu$$

This reintroduces a kinetic term for the longitudinal component $\partial_\mu\chi$ above. Additionally, the gauge invariance is broken.

My question is:

Without gauge invariance, how can we show in this case that the massive field propagates 3 degrees of freedom and not 4?


The reason why there are only 3 physical degrees of freedom, can be deduced by simply computing the Euler-Lagrange equation for the field $V^{\mu}$.
The Lagrangian reads: $$L_{V} = -\frac{1}{4} V_{\mu\nu}V^{\mu\nu} + \frac{1}{2} m^{2}V^2 $$ and the E-L equations are: $$(\Box + m^{2})V^{\sigma} - \partial^{\sigma} \partial_{\rho}V^{\rho}=0$$ Up to now, there are 4 physical degrees of freedom (i.e. the 4 components of $V^{\mu}$), but if you derive both members of the E-L equations by $\partial_{\sigma}$ you will find: $$ \Box(\partial_{\sigma} V^{\sigma}) + m^{2}(\partial_{\sigma} V^{\sigma}) -\Box(\partial_{\rho} V^{\rho}) =0 \rightarrow m^{2} (\partial_{\sigma} V^{\sigma})=0$$ If you suppose that you are describing a massive field theory (i.e. $m$ different from 0), the only way to satisy the above constrain is that $\partial_{\sigma}V^{\sigma}=0$.
At this point you can rewrite the equation of motion of your field theory as: $$(\Box + m^{2})V^{\mu}=0$$ $$ \partial_{\mu}V^{\mu}=0$$ That is a Klein-Gordon equation plus a constrain. The presence of this constrain, clearly means that you have 4-1=3 physical degrees of freedom of your theory. The latters are the 3 spatial components of the vector field $V^{\mu}$.

  • 1
    $\begingroup$ So basically, adding a mass term to the Maxwell lagrangian automatically picks the Lorentz gauge? That's pretty neat. $\endgroup$ – saolof Jun 19 '19 at 21:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.